long-text-token-classification icon indicating copy to clipboard operation
long-text-token-classification copied to clipboard

How to train with multiple gpus?

Open Tonyboy999 opened this issue 3 years ago • 5 comments

It will raise an error when I tried to train tez model on multiple gpus. model = nn.DataParallel(model, device_ids=[0,1,2,3]) AttributeError: 'DataParallel' object has no attribute 'fit'

How can I do this? Thank you so much

Tonyboy999 avatar Jan 24 '22 13:01 Tonyboy999

dataparallel doesnt work with Tez for now

abhishekkrthakur avatar Jan 24 '22 15:01 abhishekkrthakur

Are there any ways?

Tonyboy999 avatar Jan 24 '22 15:01 Tonyboy999

write your own training/evaluation loop

abhishekkrthakur avatar Jan 24 '22 15:01 abhishekkrthakur

OK, thank you

Tonyboy999 avatar Jan 24 '22 15:01 Tonyboy999

OK, thank you “write your own training/evaluation loop” how can i do this loop?thank you so much!

a2961656123 avatar Feb 13 '22 01:02 a2961656123